> Creating new safety and security standards for AI, including measures that require AI companies to share safety test results with the federal government

This will be...interesting to see how it's "tested" for safety / security.